• SONAR
  • Instability after recent Windows 10 updates - please respond Cakewalk with latest (p.3)
2016/12/18 19:24:00
Anderton
microapp
 I don't think it is unreasonable to expect at least a general time frame estimate for this to be resolved. Something like 'we expect a solution in the 2016.12 release' or 'in the 2017.01 preview'.



It's not unreasonable, but they might not know when they'll have a solution, or how many plug-ins it affects. If they estimate too far out, people will be upset it's going to take that long even though it might not. If they miss an estimate, people will be upset they were late. Unless they have a pretty firm idea of a date, it's a lose-lose situation. I'm sure they're working on it as fast as they can.
 
I've never seen (for example) NI give expected dates for a fix when a Mac OS update breaks something, probably for the same reason. However, they do warn their users not to update.
2016/12/18 19:53:34
John
I think this is limited to fairly recently created or updated plugins from CW. I think it could be the compiler used. I found out that Studio One 3 crashed when trying to scan CA2A. It also had trouble with Rapture pro and Rapture Session.  
 
No other plugin from CW have this problem. Also no other plugin from third parties seem to have any problem.
2016/12/19 01:51:06
The Grim
studio one 3 pro scans the above mentioned plugs without issue for me, but i am running windows 10 which was last updated in july.
2016/12/19 03:16:19
dlesaux
John
No other plugin from CW have this problem. Also no other plugin from third parties seem to have any problem.


That's not true in my case. I've been using Kontakt for years now with no issues but this recent spat of problems is causing stuttering, buzzing and audio engine dropouts.
2016/12/19 04:02:43
microapp
With the info available, it is hard to say whether the plugin issue (thread init) is causing other reported problems with 2016.11. It is certainly possible since much of Sonar is multi-threaded. There certainly seems to be more issues with .11 than most of the previous updates. According to Cake, the Sonar core is not affected but we will likely have to wait until the plugin issue is sorted to see if that also fixes other core issues.
2016/12/19 12:32:26
jimo
John
 
No other plugin from CW have this problem. Also no other plugin from third parties seem to have any problem.




I am also having problems with some Waves plugins, I initially noticed it on the latest version of the API 2500 (mono)
 
2016/12/19 14:01:18
abacab
I wonder if it's only happening with certain CPU configurations?  Such as i3 vs. i5 vs. i7, or different generations of Intel core, 2nd, 3rd, 4th gen ... etc.  Or with hyper-threading enabled or disabled in the BIOS? I could speculate that this may offer timing variances that are then exposed by the OS. 
 
Or perhaps Sonar plugin load balancing enabled, in come combination with the above.
 
I ask this merely because of bitflipper's earlier observations about running ProcMon.  If he started ProcMon first, then the problem did not occur.
 
I have not seen any of this problem here.  Platinum is rock steady on 2016.11, and I can insert CA-2A without a hitch.  FYI, I have kept plugin load balancing unchecked (disabled).  Not gonna rock the boat, LOL!  Running audio with ASIO at 44.1/24, 256 samples buffer.
2016/12/19 14:41:07
microapp
I too was wondering why some are affected and others not. My laptop is not affected even with CA-2A. I have postponed 2016.11/12 update on my DAW PC. I am waiting for the dust to settle but I would do it if I thought it would help ident the problem.
2016/12/19 15:40:04
bitflipper
The crash occurs on both my desktop (i7) and laptop (i5), both running the latest build of Win 10. ASIO on the desktop, WASAPI on the laptop. So those variables don't t appear to have a correlation.
 
I have not experienced any issues with Kontakt performance. Which is not to say Daniel's experience can't be related, just that it's not universal.
 
I suspect the reason that the problem didn't occur with ProcMon running is that PM intercepts system calls and logs them, adding a small amount of latency. If the issue involves a race problem in which one thread fails because it depends on another thread that hasn't had time to initialize, adding a time delay might prevent the race condition. Inserting short delays is a common practice for addressing such problems, both in software and hardware.
 
For now, I am content with my workaround: load SONAR, start a new project and then close it. A minor inconvenience.
 
 
2016/12/20 05:21:43
dlesaux
I tried creating a simple project from scratch with no FX and simply ezDrummer and instances of Kontakt to see how far how I could take this before instability issues. What I did start noticing was timing issues. I added a track with sixteenth note strings and they were lagging behind the rest of the track. My CPU usage was at about 10%. I've heard about the fact that Microsoft changed something to so with timing in their last update. Is this a manifestation of that issue?
 
I was hoping to hear something from Cakewalk regarding an explanation or timeline but nothing so far.  I also understand the fact they want to control the message out in the marketplace. The competition may want to pounce on this. 
 
I just want to get back to making music, that's all. We all do! I never used to notice Sonar, it was transparent and seamless. Now I'm constantly expecting for something to go wrong. I realize Microsoft may have really screwed up here and it may not be Cakewalk's fault..
© 2025 APG vNext Commercial Version 5.1

Use My Existing Forum Account

Use My Social Media Account